Involvement of cell surface glycosaminoglycans in chebulagic acid's and punicalagin's antiviral activities against Coxsackievirus A16 infection

Coxsackievirus A16 (CVA16) is responsible for several recent outbreaks of Hand, Foot, and Mouth Disease in the Asia-Pacific region, and there are currently no vaccines or specific treatments available. We have previously identified two tannins, chebulagic acid (CHLA) and punicalagin (PUG), as efficient entry inhibitors against multiple viruses known to engage cell surface glycosaminoglycans (GAGs). Interestingly, these two phytochemicals could also block enterovirus infection by directly inactivating CVA16 virions, which were recently reported to utilize GAGs to mediate its entry. The aim of this study is to evaluate the involvement of GAGs in the anti-CVA16 activities of CHLA and PUG. To explore a potential mechanistic link, the role of GAGs in promoting CVA16 entry was first confirmed by treating human rhabdomyosarcoma (RD) cells with soluble heparin or GAG lyases including heparinase and chondroitinase. We then performed a combination treatment of CHLA or PUG with the GAG interaction inhibitors to assess whether CHLA's and PUG's anti-CVA16 activities were related to GAG competition. Molecular docking and surface plasmon resonance (SPR) were conducted to analyze the interactions between CHLA, PUG, and CVA16 capsid. Lastly, CRISPR/Cas9 knockout (KO) of the Exostosin glycosyltransferase 1 (EXT1) gene, which encodes a transmembrane glycosyltransferase involved in heparan sulfate biosynthesis, was used to validate the importance of GAGs in CHLA's and PUG's antiviral effects. Intriguingly, combining GAG inhibition via heparin/GAG lyases treatments with CHLA and PUG revealed that their inhibitory activities against CVA16 infection were overlapping. Further molecular docking analysis indicated that the predicted binding sites of CHLA and PUG on the CVA16 capsid are in proximity to the putative residues recognized for GAG interaction, thus pointing to potential interference with the CVA16-GAG association. SPR analysis also confirmed the direct binding of CHLA and PUG to CVA16 capsid. Finally, RD cells with EXT1 KO decreased CHLA's and PUG's antiviral effect on CVA16 infection. Altogether, our results suggest that CHLA and PUG bind to CVA16 capsid and prevent the virus' interaction with heparan sulfate and chondroitin sulfate for its entry. This study provides mechanistic insight into the antiviral activity of CHLA and PUG against CVA16, which may be helpful for the development of antiviral strategies against the enterovirus.